The convenience of radio anywhere

“Our recently completed Techsurvey8 says a lot about the voracious appetite for all things mobile – especially apps,” said Fred Jacobs, President of Jacobs Media/jacAPPS. “We surveyed 57,300+ radio listeners across North America, more than half of whom now own a smartphone. Of those who download apps, over 70% download radio apps. When we asked our respondents which type of radio app they prefer, more than half opted for individually branded apps which allow for more features and customization. This is where radio is going.”

Indeed, turning a smartphone or tablet into a radio is a simple as downloading an app and streaming audio over either WiFi or a mobile data network. Which of those network options is being used provides additional insight into listening patterns. One feature of the mobile app analytics offered by Localytics is reporting on which data networks apps are used. Reviewing aggregate radio app usage in the US for about 500 apps measured by Localytics, we found that 42% of sessions were over non-mobile internet providers. In these cases, listeners were most likely using a WiFi network at home, work or at the local Starbucks.

The other 58% of radio app sessions were over mobile (cellular) data networks. Mobile network streaming can be and is still done at home, but it also provides the ability to listen “on the go,” whether the listener is walking, biking, driving or commuting. Supporting these results, Arbitron recently reported that 17% of mobile phone owners have listened to streaming radio in their cars by connecting their phones to the car stereo.

“We’re always looking for new ways to engage our listeners, and apps provide us with a great one,” says Tim Murphy of Entercom, one of the five largest radio broadcasting companies in the United States. “The convenience of being able to listen to us in the car, on the train, at home or in the office on a device you’re already carrying is great, both for us and for our listeners.”

While there are data caps to be wary of, at a streaming quality of 64 kbps a listener can stream 70 hours of radio before hitting the 2 GB monthly data plans now common in the US. By comparison, Pandora reported average listening time of 18 hours per month.

Engaging both local listeners and those beyond signal range

According to a survey by Mark Kassof and Company, most radio listeners are satisfied with their primary station. Taking away the limiting factor of a transmitter by introducing a smartphone or tablet app provides listeners with the station they are most satisfied and connected with no matter where they go, and provides stations with broader access to listeners who already have a relationship with them. Another Localytics app analytics feature that is valuable to such providers is location analysis, providing real-time metrics on where their apps are being used. Looking at aggregate radio streaming data over WiFi which can be used to more accurately determine metro area location, we found that 46% of users listened in areas that are outside terrestrial broadcast range.

This also means that more than half of terrestrial radio app listening is taking place in locations within the radio stations’ traditional broadcast reach. Even within a station’s range, offering a smartphone app gives listeners the convenience of listening when signal coverage is spotty, or where a standalone radio might be inconvenient. Radio stations with associated smartphone apps can give their listeners a more varied and flexible experience both within and outside of their normal broadcasting region.

Analyzing and optimizing radio apps

The nature of mobile apps also gives radio stations unprecedented insight into listening preferences and what content is driving engagement. With proper app analytics, broadcasting companies can take best practices from their top apps – design elements, content, ad placement, etc. – and apply them to the rest of their radio stations. With Localytics’ app analytics platform implemented, station management can see exactly how long users are listening, what they’re listening to most often, and how loyal they are. This understanding of what users are doing and the ability to quickly pivot their strategy has the radio industry excited about mobile apps as another way to engage with their listeners.

“Radio listenership is stronger than ever, and with streaming apps we can give our listeners another level of convenience and access to their favorite stations,” says Murphy. “Both we as broadcasters and our listeners benefit from mobile radio apps. We will continue to take advantage of this new medium, and we’re very excited about what we can do.”

Best practices example: Filtering regional editions

Let’s take the example of a publisher that started in the US, but recently launched a new international edition of a popular magazine. The publisher wants to analyze audience and readership of their magazine both globally and by regional edition.

To do this, the publishers simply uses the same Localytics app key (similar to a profile in web analytics) and captures the magazine edition as a custom dimension. Each usage, engagement, location map and event analysis can then be filtered by edition. In the following example, the Article View event is first split by Subscriber type (annual, monthly, trial) and then the results are filtered to include only the international edition.

Getting started with Filters

Premium and Enterprise automatically have access to the new filters feature. Simply log in to your Localytics app analytics dashboard, choose an app, click a usage, engagement, events or map chart and you’ll see the new Apply Filters option next to the date selector. The analysis will update as soon as you apply the filter, and the filter will persist across new reports. Premium customers can apply one filter at a time. Enterprise customers can combine up to three filters.

These dimensions and their values can be used as filters:

• App version
• Carrier or operator
• Country
• Model
• New vs returning
• OS version
• Custom dimensions

Community customers want to try filters should contact us and request a free trial of Localytics Premium. Others can access a live demo of Localytics Premium.

Pew Research Center’s 2011 study showed how prevalent casual users are on news websites. Their data, provided in part by Nielsen, showed that nearly 70% of users went to a given news site only once or twice per month, and less than 10% visited 10 or more times. By comparison, our latest research show that just 29% of users launch a news app only once a month, and an impressive 24% launch it 10+ times.

News app and news website usage are very different

Much of the research by Pew’s Project for Excellence in Journalism (PEJ) focused on the “casual users,” or users who come to read a news site because of inbound traffic generation techniques – SEO, social shared links, etc. – rather than the brand. While they may be reading a story about a particular topic on the New York Times website, it’s often the case that what brought them to the site was a Google search for the topic, not their desire to read the New York Times.

With mobile apps, a user has self-identified as a fan of a particular news brand by downloading the app. Rather than coming to the news brand based on its search rankings they find stories based on the brand’s app. The nature of the app means that users have placed the power of choosing what news to display in the news organization’s hands, which means that if the app succeeds in creating a great news experience the user will return often.

Web vs. app boils down to inbound vs. retention

This does point to a very fundamental difference in analyzing the success of an online news property vs. a news app: while an online news property depends more on bringing in more traffic via SEO and social media, news app marketers should be setting user retention and loyalty goals.

An app download is a good start, but as this research shows, it’s only the beginning of an app’s success. Marketers of news apps should set goals of high sessions-per-user, long session lengths, number of articles read, optimized subscription funnels, and overall engagement time.

Custom dimensions means custom user segments

Certain dimensions are standard with Localytics – things like device, location, OS, new vs. returning and so on. Dimensions are then cross-tabulated, filtered or combined to analyze your app usage and discover actionable insights and user segments; such as what hour of the day generates the most new users and should therefore be a focus of future promotional ads.

Custom dimensions are additional dimensions that you define, specific to your application. An ideal custom dimension will help you segment app usage by a handful of values. Good examples include registration status, subscriber type and gender. They may be populated based on actions taken within the app or an existing resource, such as a customer relationship management (CRM) database.

Useful custom dimensions

Needs will vary among app verticals, but here are some ideas for useful custom dimensions:

• Demographics – age, gender, location
• Subscription status – free, trial and premium subscribers
• Purchasing history – users who never buy, bought once, bought 2-10 times
• Brick-and-mortar visits – users who used your app in one of your locations
• Multi-app usage – users who have one or multiple of your available apps
• Degree of usage – light, moderate and heavy users
• Ad clickers – users who have clicked in-app ads vs. users who never click

Best practices example: Purchasers vs. non-purchasers

In many mobile commerce apps, a smaller segment of users make a disproportionate share of the purchases. The most successful m-commerce apps will carefully analyze this valuable user segment and work to find and develop more like them.

To discover these patterns and opportunities, a custom dimension can be created to record app sessions coming from users who have a history of making purchases and those who don’t. Using these data, app usage can be segmented by different user value levels and segments like daypart, device, loyalty or other actions. These segments can paint a picture of the “typical” likely-to-purchase app user, allowing you to tailor messaging, offers and even user acquisition campaigns.

For this example, we compare user loyalty – that is, how many sessions they engage with the app – of the purchaser and non-purchaser user segments.

App loyalty of users who haven’t purchased

App loyalty of users who have purchased

As we can see, users who made a purchase are much more loyal than users who never have. The number of “purchased” users who have interacted with our m-commerce app 10 or fewer times is only 3%, compared with 26% of non-purchasing users. In other words, users tend not to buy during the first few sessions–an insight that can drive marketing efforts to keep users engaged until they are predisposed to make a purchase.

Localytics research has found that the average in-app purchaser makes their first purchase 12 days after their first session. Using your own app’s specific numbers allows you to prepare timed push notifications, in-app special offers and special coupons for users who are approaching the average first-purchase threshold. Once a user has interacted with your app enough to be considered a candidate, these timed and timely offers can give them that final push to make that initial purchase.

Getting started with custom dimensions

Custom dimensions is one of the central features of Localytics Enterprise. If you’re already an Enterprise subscriber, contact your enterprise specialist for help setting up your app’s custom dimensions. If you’re interested in becoming a Localytics Enterprise customer, contact us today and we’ll provide more details on how you can take your mobile app analytics to the next level.

See how OfficeDrop used Localytics features like screen flows and funnels to fix this problem. Learn how they tripled their user registration rate – the data they used, the decisions they made, and the ongoing improvements they’ve made to their app. Download the case study today.

73% of Android handset usage with the same OS build

Across all apps using Localytics, a full 73% of Android usage came from devices running a variant of Android 2.3. While the build, known as “Gingerbread”, is not the most recent, from a fragmentation perspective it should be good news to developers that such a large majority of users are running the same Android OS version.

Add Android 2.2 “Froyo” to the mix, and the majority becomes even more convincing – 23% of Android user sessions were running some flavor of “Froyo”. Between the two, Android developers can be confident that they only need to actively target two Android OS builds in order to achieve 96% compatibility with the Android ecosystem.

Most popular Android handset size is 4.3 inches

Another concern for Android developers is screen size and resolution. Of all app usage analyzed for this study, 41% of all sessions came from Android devices with 4.3 inch screens, by far the most popular size. 4 inch screens accounted for 22% of sessions, 3.2 inch screens for 11%, and 3.7 inch screens contributed 9%.

Resolutions were even less fragmented, however, with the most widely-seen screen resolution – 800 x 480 pixels – contributing 62% of the study’s sessions. The next most popular screen resolutions were 480 x 320 (14%), 960 x 540 (6%), 480 x 854 (5%) and 320 x 240 (5%).

For both screen size and resolution, Android developers have more to deal with than iOS developers, thanks to Apple’s single handset form factor. However, with five options accounting for more than 90% of all Android app usage, the fragmentation is not particularly daunting.

Android tablets showing similar patterns, Kindle Fire and Nook Tablet leading the way

For developers working on Android tablet applications, the fragmentation numbers are similar to those seen in handsets. Led by Amazon’s Kindle Fire, Barnes and Noble’s Nook and Samsung’s Galaxy Tab – the three most widely-used Android tablets as measured by Localytics – nearly three quarters of all Android tablet usage is from devices with the same specs.

Screen resolution and size are actually even less fragmented than handsets – 74% of Android tablet usage takes place on 7 inch devices with 1024 x 600 resolution. 22% are 10.1 inch devices with 1280 x 800 resolutions, so by taking into account two screen size/resolution combinations, developers should be able to easily reach nearly all of the Android tablet market.

OS distribution is similarly stable, with 71% of Android tablets running Android 2.3 “Gingerbread”. Most of the remaining tablets are running some flavor of Android 3 “Honeycomb” – 14% are powered by Android 3.2, 7% by 3.1. The rest run “Froyo”.

Why Android developers should not be as worried about fragmentation

In all, while Android developers do have to think about an element of fragmentation foreign to iOS developers, the problem seems to be much smaller than made out to be. Android devices, particularly tablets, tend to be very similar in OS version, screen size and screen resolution. Although iOS developers only have two form factors to craft their applications for, the similarities of Android’s most-used devices should make it less of a challenge to develop for than often thought.

Methodology

For the purposes of this study, Localytics looked at the number of user sessions seen by apps running Localytics as their analytics provider. All data is usage-based. The time frame is the two-week period ending Jan. 22, 2012.

For All Users

Set default time zone for reports

Customers can now set the default time zone for all reports in the dashboard. To change your apps’ time zones, simply log into your Localytics dashboard, click on the Administration tab, and hit the Edit button on the applicable app. You will now find that, in addition to the app’s name and category, a drop-down menu for time zone is now available.

Improved session length analyses

Get even better insight into your users’ session lengths – your session length metrics now automatically filter out obviously erroneous session lengths (below zero seconds and above 12 hours). In addition to average session length, we now also report median session length.

For Premium and Enterprise

Monthly cohorts splits added to all charts

User-based cohort splits are now available in the Usage and Event charts, with cohorts determined by the time that users first launched your app. This gives another view of retention by splitting session, location, features and more by the month in which users were acquired. For example, view sessions by country and first month the app was used.

Weekly cohorts are also available with Advanced usage reports. Enterprise customers can add daily cohorts as a dimension in Query Builder.

New dimensions for Advanced Reporting and Query Builder

Premium and Enterprise customers can create and save custom charges by cross-tabbing any two dimensions. Query Builder goes a step further and gives Enterprise customers ad hoc, cross-app queries combining up to four dimensions through a simple web UI, with the results download directly to their computer. Three additional dimensions have been added to both the Advanced reports and Query Builder:

• Language (Locale)
• Daily, weekly and monthly cohort splits
• Jalibroken

For Enterprise

Custom dimensions

Custom dimensions are now available in Localytics dashboard charts. By creating custom dimensions you can segment your analyses with user profile data that specifically meet your needs, including gender, subscriber status, and purchaser vs. non-purchaser. Cross-tab your custom dimensions with other standard dimensions (most popular hour of the day for purchasers vs. non-purchasers, for example) and discover actionable insights into app usage.

For help getting your custom dimensions implemented, please contact Localytics support today.

In research from last year, Localytics found that 26% of apps are used only once after being downloaded. As a result, the app publisher community has made a concerted effort to shift away from incentivized downloads and towards engagement and overall customer lifetime value (CLV).

With this new data, that shift is shown to be even more important. With a high correlation between highly engaged users and in-app purchases, developers and publishers with a goal of maximizing CLV have even more reason to obsess over user retention. The more your users interact with your app, the more likely you are to get a chunk of the $5.6 billion in in-app purchases expected by 2015.

Engagement leads to higher CLV

Even more, the users who wait and interact with an app multiple times before making their first in-app purchase are more valuable in the long run, making 25% more in-app purchases over their lifetime as a customer. First session purchasers will make an average of 2.8 purchases in a given app during their user lifetime, compared with 3.5 purchases for all other purchasers.

While moving users towards a purchase as quickly as possible is often considered the primary objective, this data suggests that turning purchasers into loyal, repeat users should be a top priority. While a first-session purchase is an excellent result, our data found that only 16% of users who make a purchase their first app session will go on to engage with that app 10 or more times, compared with 26% of overall app users. Thus, it is imperative that app developers and marketers pay special attention to their analytics solution’s loyalty features.

Grow customer lifetime value

By building your app’s engagement and your brand’s presence in a user’s mind, you can generate better overall revenue based on a loyal base of repeat users. Given the 12 day average time between downloading an app and making a purchase, driving loyalty across a period of weeks will often generate greater revenue.

At Localytics, we help app developers and marketers build more successful and profitable apps with the highest possible CLV through retention analysis and using funnel analysis to optimize in-app purchases. To enable these analyses, app publishers track each purchase and the steps taken to complete the purchase. For the purposes of this study, Localytics analyzed nearly 30 million users’ in-app purchasing to extract patterns publishers can use to benchmark their own apps’ performance.

Apps Expand and Evolve

Apps for all those smart phones continue to be insanely popular with Apple reporting 18 billion downloads for iOS and Google reporting 10 billion downloads for Android. Apps also expanded far beyond their “mobile” heritage. iPads and other tablets may still be considered mobile, but many rarely leave the house. Then there are smart TVs, smart appliances and whatever else we’ll see emerge at consumer electronics shows next year.

The way apps are built and monetized continues to evolve too. In-app purchases and subscriptions are now very popular and successful additions to free, paid and ad-supported models. App publishers are also exploring new development languages and distribution models by adding HTML5 to the already popular Objective-C and Java.

We won’t offer yet another list of predictions for 2012, but we’re more excited than ever about the opportunities for apps and the role Localytics can play.

Thank You and Looking Forward to Next Year

All of us at Localytics want to thank our customers and partners for helping to push our business forward in 2011.

Cheers to an incredible 2011 and an even better 2012!